The present invention relates to a road obstacle monitoring device and more in detail to a road obstacle monitoring device, in which road obstacles such as wave flooding, land slip, etc. are detected by means of a monitoring TV camera; detected image signals are subjected to image processing and it is judged whether there are road obstacles or not; and if there are road obstacles, a warning signal is outputted immediately in order to be able to take measures at sites against the road obstacles and to give warning to drivers of passing vehicles, etc.
Heretofore, occurrence of this kind of road obstacles is verified often by a monitoring patrol by monitors or by information by passengers, drivers of passing vehicles, etc., who are going to pass through a place of the occurrence, after the road obstacles have really occurred. On the other hand, depending on the location of the road, suitable detecting sensors are set before hand to verify the road obstacles by measuring outputs from the detecting sensors.
Consequently, by the prior art means for detecting the road obstacles as described above, in the case where the road obstacles take place, they cannot be verified, until a monitor goes to the place of occurrence for the monitoring or an informer informs him of the occurrence of the road obstacles, which causes delay of judgment to be done for securing the safety for the traffic. Further, depending on the kind of detecting sensors, there is an undesirable problem that an extremely large number of detecting sensors are required for effecting measurements and detections of unspecified road locations and extents of road obstacles.